1. Field of Invention
This invention relates to an assay system designed to detect a protein biomarker in urine that is diagnostic for interstitial cystitis (IC). The cytoskeletal associated protein 4 (CKAP4) is subcloned in frame with yellow-fluorescent protein (YFP) creating a fusion protein (CKAP4-YFP). CKAP4-YFP is ectopically expressed in the human cervical carcinoma cell line, HeLa, at the cell surface, creating a cell-based system. Urine samples from patients who exhibit symptoms consistent with IC are added to the cell-based system. The presence of a 9 amino acid glycopeptide, antiproliferative factor (APF), in urine is unique to patients with IC. APF specifically binds to CKAP4-YFP at the cell surface. Binding of APF to CKAP4-YFP enables transit of CKAP4-YFP-APF to the nucleus. Transit of this protein complex to the nucleus is measured using high-throughput immunofluorescence microscopy. Nuclear localization of the YFP signal is positive for the presence of APF in urine and diagnostic for IC. The cell-based diagnostic system is a significant and surprising advance in diagnosis of IC and has commercial applications relevant to greater than 147/100,000 women and 41/100,000 men who suffer from symptoms consistent with IC.
2. Description of Related Art
Interstitial cystitis/painful bladder syndrome (IC/PBS) is a poorly understood, chronic, urinary bladder disease manifested by bladder pain and increased frequency and urgency of urination (Sant and Hanno, 2001). IC has no single, definable presentation, but is best viewed as a continuum extending across decades of an individual's life, beginning with mild, intermittent symptoms that become more severe and constant over time. For many, pain is the most dominating and debilitating symptom, becoming so severe as the disease progresses that it significantly affects patients' quality of life (Marshall, 2003). As many as 1.3 million American men, women, and children of all ages and races have IC; however, the disease is far more common in women (Clemens et al, 2007). Approximately 60/100,000 adult American women are affected, with more recent studies suggesting a substantially higher prevalence (Curhan et al, 1999; Leppilahti et al, 2005). Indeed, the prevalence in the managed care population was 197/100,000 for women and 41/100,000 for men when diagnosed using ICD-9 code (Clemens et al, 2005). Because the causes of IC are unknown, current treatments are aimed only at providing symptom relief. Many patients have noted an improvement in symptoms after bladder distention has been performed to diagnose the condition; however, researchers are not sure why this helps. Bladder instillation (also called a bladder wash or bath) with dimethyl sulfoxide given every week or two for 6 to 8 weeks has also been noted to improve symptoms temporarily. Because DMSO passes into the bladder wall during this procedure, physicians believe that it may reach tissue more effectively to reduce inflammation, block pain, and also prevent muscle contractions that cause pain, frequency, and urgency. Pentosan polysulfate sodium (the first oral drug developed for IC; Elmiron) given orally (100 mg three times a day) has improved symptoms in 30% of patients. Bacillus Calmette-Guerin (BCG), a common intravesicular agent used in bladder cancer, has also been shown to improve symptoms (Peters et al, 1998). Electrical nerve stimulation and lifestyle interventions, including dietary modifications, bladder training, and exercise, are other approaches to symptom relief. Surgical procedures such as fulguration and resection of ulcers, bladder augmentation, or bladder removal (cystectomy) are a consideration only if all available treatments have failed and if the pain is disabling; however, most surgeons are reluctant to operate because some patients still have symptoms after surgery.
IC presents with symptoms that are identical to those detected in patients with bacterial infections and other disorders of the urogenital tract (Table 1) (Wein et al, 1990). In addition to the diseases listed in Table 1, differential diagnosis includes several other possibilities, including overactive bladder (although these patients do not have bladder pain), chronic urinary tract infection in an otherwise normal urinary tract, vulvodynia (or tender vulva) which may coexist with IC, and endometriosis which may also coexist with IC. These conditions are individually treated using different clinical approaches. Significantly, there is no commercially available diagnostic test for IC, making it difficult to distinguish it from other conditions with similar symptomology and difficult to develop an appropriate treatment strategy. These gaps in knowledge directly affect patient care: in the past, patients experienced an average lag time of five to seven years before they received a diagnosis of IC (Curhan et al, 1999). With current prevalence estimates for IC in the managed care population of 197/100,000 for women and 41/100,000 for men (Clemens et al, 2005), hundreds of thousands of people would have to be screened in order to differentiate IC from other conditions (Clemens et al, 2005). At present the only available diagnostic specific to IC is an ulcer of the bladder mucosa, which is called Hunner's ulcer. Unfortunately, this condition is not very helpful for the purposes of diagnosis because it is found in 10% or fewer IC cases (Sant and Hanno, 2001). Thus, there is a critical need for improved strategies to accurately diagnose IC.
TABLE 1Diseases that may mimic symptoms of IC(Wein et al, 1990)Active genital herpesVaginitisBladder stonesBladder cancerCancer of the uterus, cervix, vagina, or urethraUrethral diverticulumRadiation cystitisCyclophosphamide cystitisBladder tuberculosisNeurogenic bladderDrug effects: aspirin, NSAIDs, allopurinol
Etiology of Interstitial Cystitis: Despite multiple hypotheses about the primary cause of IC, the underlying molecular mechanism of this disease remains completely undefined. Active theories include increased permeability of the bladder mucosa, abnormal neuronal function, mast cell activation, autoimmunity, infections, and toxic or antiproliferative substances in the urine (Erickson et al, 1999); however, insufficient data exist to definitively establish their roles in the pathology of IC. One of the most consistent findings in bladder mucosal biopsies from IC patients has proven to be thinning or erosion of the bladder epithelium (Keay, 2008; Leiby et al, 2007). It is speculated that normal repair of the damaged bladder epithelium does not occur in patients who develop IC; thus, urine contents, such as potassium, can leak into the bladder interstitium, possibly leading to mast cell activation and histamine release. In response, C-fiber nerves become activated, triggering the release of Substance P as well as immunogenic and allergic responses. These events are thought to lead to progressive bladder injury, which may promote spinal cord changes in some patients, causing chronic neuropathic pain (Hanno, 2007). Research studies have demonstrated abnormal cell signaling in IC bladder epithelial cells as compared to controls. The expression of several proteins, including group D-related human leukocyte antigen (HLA-DR) and inter-cellular adhesion molecule 1 ICAM-1, the secreted proteins IL-1, TNFα, and various epithelial growth factors (i.e., HB-EGF, EGF, and IGF1) have been shown to have altered expression in bladder biopsies and/or explanted primary bladder epithelial cells from IC patients (Keay, 2008). In addition, increased paracellular permeability and abnormalities in tight junction protein expression (i.e, zonula occludens-1 [ZO-1], occludin, and claudin 1, 4, and 8) have been demonstrated in explanted bladder epithelial cells grown for IC patient biopsies as compared to cells from normal controls (Zhang et al, 2005; Keay et al, 2003a; Zhang et al, 2007). These abnormalities may be related to altered bladder epithelial cell differentiation in IC patients, with increased E-cadherin expression and decreased expression of ZO-1 and uroplakin III, and altered cytokeratin gene expression in vivo and/or in vitro (Keay et al, 2003a; Zhang et al, 2007; Hauser et al, 2008; Southgate et al, 2007; Slobodov et al, 2004; Laguna et al, 2006; Keay, 2008). Bladder epithelial cells from IC patients also exhibit profoundly decreased proliferation (Keay et al, 2003b), decreased expression of cyclin D1 and JNK (Keay et al, 2003a), and increased paracellular permeability in vitro in the absence of exogenous serum or growth factors as compared to cells from normal controls (Zhang et al, 2005). Collectively, these findings provide additional indirect evidence for altered epithelial cell signaling in IC.
Current Approaches to IC Diagnosis: The diagnosis of IC in the general population is based on the presence of pain related to the bladder, usually accompanied by frequency and urgency to urinate, in the absence of other diseases that could cause the symptoms. Diagnostic tests that help rule out other disease include urinalysis, urine culture, cystoscopy, biopsy of the bladder wall, distention of the bladder under anesthesia, urine cytology, and laboratory examination of prostate secretions (Sant and Hanno, 2001). Multiple urine markers for IC have been described including antiproliferative factor (APF), EGF, IGF binding protein-3, and interleukin (IL)-6, which are significantly increased in IC and HB-EGF, cyclic guanosine monophosphate, and methylhistamine, which are significantly decreased in IC. In a study that compared these markers, APF was found to have the least overlap in the IC and control groups (Erickson et al, 2002). APF was originally purified from the urine of IC patients, and bioactivities attributed to APF include suppression of urothelial cell proliferation; increases in transcellular permeability; lowering of the expression of proteins that form intercellular junctional complexes; and reduction in the production of HB-EGF from urothelial cells (Keay et al, 2004). APF activity is detectable in the urine of approximately 95-97% of IC patients who fulfill the symptomatic, exclusionary, and cystoscopic NIDDK criteria for IC (Keay et al, 2001; Keay et al, 1996) and 94% of patients who fulfill symptomatic and exclusionary criteria alone (Keay et al, 2007). The specificity of APF for urine from IC patients (vs. normal controls or patients with a variety of other urogenital disorders (Keay et al, 2001)) suggests that it is useful as a diagnostic marker for IC.
Potential Role of APF Interstitial Cystitis: APF is a low molecular weight sialoglycopeptide secreted specifically from bladder epithelial cells in patients suffering from IC (Keay et al, 2004; Keay et al, 2000). The peptide sequence of APF is identical to residues 541-549 of the 6th transmembrane domain of Frizzled 8, a Wnt ligand receptor. The glycosyl moiety of APF consists of sialic acid α-2,3 linked to galactose P1-3-N-acetylgalactosamine, which is α-O-linked to the N-terminal threonine residue of the nonapeptide. APF has been shown to profoundly inhibit the proliferation of normal bladder epithelial, bladder carcinoma, and cervical adenocarcinoma cells in vitro (Keay et al, 2004; Keay et al, 2000). Furthermore, APF can induce multiple changes in the pattern of cellular gene expression including decreased production of HB-EGF and increased production of E-cadherin, resulting in a more differentiated bladder epithelial cell phenotype (Keay et al, 2000; Keay et al, 2003). APF was also recently determined to decrease tight junction protein (zonula occludens-1 and occludin) production and increase paracellular permeability of normal bladder epithelial cell monolayers similar to changes seen in cells from patients with IC in vitro (Zhang et al, 2005). This accumulation in urine of a bioactive factor, capable of altering the behavior of urothelial cells, is consistent with the clinical observation of epithelial thinning and denudation observed in IC bladder tissue.
The potency of APF (EC50 in the picomolar range), its varied effects on bladder epithelial cell protein expression and proliferation, and the requirement for a hexosamine-galactose disaccharide linked in a specific alpha configuration to the backbone peptide for activity, all indicate that APF's effects are mediated by binding to and activating a receptor.
In 2006, Conrads et al identified cytoskeletal associated protein 4 (CKAP4) (also known as p63, CLIMP-63, ERGIC-63), as a high affinity receptor for APF (Conrads et al, 2006). CKAP4 has also been identified as a functional cell surface receptor for tissue plasminogen activator (tPA) in smooth muscle cells (Razzaq et al, 2006) and for surfactant protein A (SP-A) in rat type II pneumocytes (Gupta et al, 2006). CKAP4 is a nonglycosylated, reversibly palmitoylated, type II transmembrane protein. At its NH2 terminus, CKAP4 has a 106-amino acid long cytosolic tail, a single transmembrane domain, and a large extracytoplasmic domain of 474 amino acids. Original studies of CKAP4 described it as an endoplasmic reticulum (ER) resident protein (Schweizer et al, 1994). Subsequently, it was shown to aid in the anchoring of rough ER to microtubules in epithelial cells (ie, COS and HeLa) (Klopfenstein et al, 1998). This function requires a direct interaction between the cytoplasmic N-terminal tail of the protein to microtubules and is regulated by phosphorylation (Vedrenne et al, 2005).
A system for detecting APF bioactivity utilizes 3H-thymidine incorporation to measure the cellular proliferation of cultured normal bladder epithelial cells (explanted from human bladder tissue) following exposure of the cells to patient urine. This bioassay does not discriminate from other factors present in urine that have also been shown to regulate cellular proliferation (e.g., heparin binding epidermal growth factor-like growth factor (HB-EGF), epidermal growth factor (EGF), interleukin-1 (IL-1) and insulin-like growth factor 1 (IGF1), etc.) A system for detecting APF by generating specific antibodies against this 9 amino acid glycopeptides is problematic. It's size of 9 amino acids indicate it is not a good immunogen as this is the minimal size epitope for any antibody to recognize and well below minimal sized polypeptides used to generate specific antibodies. Further, developing a specific antibody against APF is also problematic given that its 9 amino acids are identical to sequences found in a receptor protein called Frizzled 8, presenting the potential to cross-react with a larger protein that shares this sequence. A second system for detecting APF is the focus of the present invention, includes measuring its presence based on bioactivity, a cell-based system that expresses a specific receptor for APF and upon presentation of urine samples to the cell based system, binding of APF induces translocation of the receptor to the nucleus. The receptor protein is fused to a fluorescent indicator signaling protein. The presence of the fluorescent signal in the nucleus is quantified by high-throughput microscopy. Translocation of the fluorescent signal into the nucleus is diagnostic for the presence of APF and for IC.
In addition, Palmitoylation is the posttranslational addition of the 16-carbon palmitate group to specific cysteine residues of proteins (Smotrys and Linder, 2004) via a labile thioester bond. Unlike other forms of lipidation, such as myristoylation and prenylation, palmitoylation is reversible which allows for dynamic regulation of protein-membrane interactions, trafficking between membrane compartments (Wedegaertner and Bourne, 1994; Jones et al., 1997; Moran et al., 2001; Zacharias et al., 2002), and synaptic plasticity (el-Husseini Ael and Bredt, 2002). For many years it was believed that palmitoylation occurred primarily by autocatalytic mechanisms (Bizzozero et al., 1987; Bano et al., 1998); however, the recent discovery of a family of palmitoyl acyl transferase (PAT) enzymes that catalyze protein palmitoylation has reversed this notion, expanding the complexity of the mechanisms by which palmitoylation is regulated (Lobo et al., 2002; Roth et al., 2002; Fukata et al., 2004; Linder and Deschenes, 2007).
PATs are encoded by the ZDHHC gene family and are characterized by an Asp-His-His-Cys motif (DHHC) within a cysteine-rich domain (CRD). The DHHC and CRD domains are essential for palmitoyl acyl transferase activity (Roth et al., 2002; Fukata et al., 2004; Sharma et al., 2008). Twenty-three genes encoding proteins with DHHC-CRD domains have been identified in mouse and human databases (Fukata et al., 2004). Of these, at least seven have been shown to be associated with human disease: DHHC8 with schizophrenia (Mukai et al., 2004); DHHC 17/HIP 14 with Huntington's disease (Yanai et al., 2006); DHHC15 and DHHC9 with X-linked mental retardation (Mansouri et al., 2005; Raymond et al., 2007); and DHHC2, DHHC9, DHHC17, and DHHC11 with cancer (Oyama et al., 2000; Ducker et al., 2004; Mansilla et al., 2007; Yamamoto et al., 2007). In many of these examples, the absence of PAT expression and subsequent failure to palmitoylate target substrates is the underlying problem.
Although now recognized as a PAT, DHHC2 was previously known as ream for reduced expression associated with metastasis. As the name suggests, this gene was first identified because its expression level was consistently and significantly reduced in clonal murine colorectal adenocarcinoma cell lines with high metastatic potential, but not in clonal lines derived from the same tumor that did not metastasize (Tsuruo et al., 1983; Oyama et al., 2000). It was concluded that ream expression is inversely related to the metastatic potential of a cell, leading to speculation that this gene normally suppresses one or more of the processes by which cancer cells escape from blood vessels, invade into and proliferate in a target organ, and induce angiogenesis and form metastatic foci. Human ZDHHC2 maps to a region of chromosome 8 (p21.3-22) that is frequently deleted in many types of cancer, including colorectal (Fujiwara et al., 1993; Ichii et al., 1993; Fujiwara et al., 1994) hepatocellular carcinoma (Emi et al., 1993; Fujiwara et al., 1994), non-small cell lung (Fujiwara et al., 1993; Ohata et al., 1993), and cancers of the breast (Yaremko et al., 1996; Anbazhagan et al., 1998), urinary bladder (Knowles et al., 1993), and prostate (Bova et al., 1993). Loss of heterozygosity on chromosomal band 8p22 has been shown to be a common event in some epithelial tumors, pointing toward the likelihood that the region harbors potential tumor suppressor genes (Emi et al., 1993; Fujiwara et al., 1993; Ichii et al., 1993; Ohata et al., 1993) Because DHHC2 has no other known signaling properties beyond palmitoylation, knowledge of its target substrates in a cancer cell line could yield significant clues about its role in metastasis and tumor suppression. In previous work, we used a novel, proteomic method called PICA to identify the target substrates of DHHC2 in HeLa cells, a cervical adenocarcinoma cell line. We determined that cytoskeletal associated protein 4 (CKAP4, also known as p63, ERGIC-63, and CLIMP-63) is a principle, physiologically important substrate of DHHC2 (Zhang et al., 2008).
CKAP4 is a reversibly palmitoylated, type II transmembrane protein that has been shown to anchor rough ER to microtubules in epithelial cells (ie, COS and HeLa) (Schweizer et al., 1993a; Schweizer et al., 1993b; Schweizer et al., 1994; Schweizer et al., 1995a; Vedrenne and Hauri, 2006). This function requires a direct interaction between the cytoplasmic N-terminal tail of the protein to microtubules and is regulated by phosphorylation of three critical serine residues (Klopfenstein et al., 1998). More recently, CKAP4 has been identified as a functional cell surface receptor for antiproliferative factor (APF) (Conrads et al., 2006), a low molecular weight, Frizzled-8 protein-related sialoglycopeptide secreted from bladder epithelial cells in patients suffering from the chronic, painful bladder disorder, interstitial cystitis (IC) (Keay et al., 2000; Keay et al., 2004a). APF profoundly inhibits normal bladder epithelial cell growth (Keay et al., 1996; Keay et al., 2000; Keay et al., 2004a). APF also inhibits the proliferation of bladder carcinoma cells and HeLa cells in vitro with an IC50 of ˜1 nM (Keay et al., 2004a; Conrads et al., 2006; Keay et al., 2006). Binding of APF to CKAP4 results in inhibition of cellular proliferation and altered transcription of at least 13 genes known to be involved in the regulation of proliferation and tumorigenesis (including E-cadherin, vimentin, cyclin D1, p53 and ZO-1) (Keay et al., 2003; Conrads et al., 2006; Kim et al., 2007).
The present invention is directed to the effects of reduced CKAP4 palmitoylation on APF-mediated signaling by silencing the expression of DHHC2 with targeted siRNA. Our data show that DHHC2-mediated palmitoylation of CKAP4 is a critical event regulating CKAP4 subcellular distribution, APF-stimulated changes in cellular proliferation and gene expression, as well as APF-independent changes in cellular migration.
All references cited herein are incorporated herein by reference in their entireties.